This project continues our study of those biochemical changes that are early, critical events in the onset of virus-induced leukemia by employing AKR mice that carry an oncornavirus. A second model for leukemogenesis that we employ is a chemically induced lymphoma produced by a single injection of the antineoplastic agent, 5-(3,3-dimethyl-1-triazeno)-imidazole-4-carboxamide. Particular emphasis is placed on those changes that occur in the preleukemic state and that eventually allow the establishment of the leukemic clone. This laboratory has shown many enzymatic differences between normal thymus cells and cells of the lymphoma. On the other hand, the only detectable change in the preleukemic state, with no overt signs of leukemia, was an elevation of adenylate cyclase and cyclic AMP. We propose that the expansion of the leukemic clone in the thymus requires the turning off of the normal immune response and of the proliferation of normal lymphocytes by producing elevated levels of cyclic AMP in these cells. One mechanism that we are examining for the turning off of normal cells is suppression by glycolipids. We are studying the inhibition of thymocytes by gangliosides that appear to shed by tumor cells. Such gangliosides may mediate the observed changes in cyclic nucleotides. The study will hopefully provide a focal point for interfering with the chain of metabolic events that results in the virus-induced malignant evolution of lymphoid cells.